Update README.md

README.md

@@ -25,7 +25,7 @@ language:
 - **Model Developers:** Neural Magic
 
 Quantized version of [Meta-Llama-3.1-70B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-70B-Instruct).
-It achieves an average score of 78.69 on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) benchmark (version 1), whereas the unquantized model achieves 78.67.
+It achieves an average score of 82.78 on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) benchmark (version 1), whereas the unquantized model achieves 82.74.
 
 ### Model Optimizations
 
@@ -131,6 +131,7 @@ oneshot(
 ## Evaluation
 
 The model was evaluated on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) leaderboard tasks (version 1) with the [lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness) and the [vLLM](https://docs.vllm.ai/en/stable/) engine, using the following command:
+A modified version of ARC-C was used for evaluations, in line with Llama 3.1's prompting.
 ```
 lm_eval \
   --model vllm \
@@ -166,11 +167,11 @@ lm_eval \
   <tr>
    <td>ARC Challenge (25-shot)
    </td>
-   <td>70.65
+   <td>95.05
    </td>
-   <td>70.31
+   <td>94.88
    </td>
-   <td>99.52%
+   <td>99.82%
    </td>
   </tr>
   <tr>
@@ -216,9 +217,9 @@ lm_eval \
   <tr>
    <td><strong>Average</strong>
    </td>
-   <td><strong>78.67</strong>
+   <td><strong>82.74</strong>
    </td>
-   <td><strong>78.69</strong>
+   <td><strong>82.78</strong>
    </td>
    <td><strong>100.0%</strong>
    </td>